Chapter 2: Preclinical Development –
Alchemists, Animals, and PK/PD Models
Pfizer’s goal was ambitious.
The new compound had to accomplish several things at once.
It needed to inhibit virtually all known resistance mutations of both ALK and ROS1 while at the same time providing better control of brain metastases than the available second-generation ALK inhibitors.
____________________________________
Now it was the medicinal chemists’ turn.
Anyone who has watched Breaking Bad can probably imagine what the Pfizer research laboratories looked like at the time. Fortunately, everything taking place there was entirely legal—and instead of producing crystal meth, the scientists were trying to create a molecule capable of becoming the most potent and safest ALK and ROS1 inhibitor possible.
The “Heisenbergs” at Pfizer later summarized their work in rather different words:
“Structure-based drug design, lipophilic efficiency optimization, and optimization of physicochemical properties led to the discovery of potent macrocyclic ALK inhibitors with favorable ADME properties, low P-glycoprotein-mediated efflux, and high passive permeability. These structurally unique macrocyclic inhibitors demonstrated activity against wild-type ALK and clinically observed ALK resistance mutations. Significant synthetic challenges were overcome through the development of novel transformations that enabled the application of macrocycles to modern drug discovery.”
____________________________________
[Fig. 2 – Lorlatinib molecule structure]
____________________________________
May I introduce…
PF-06463922
—or, if you prefer its full chemical name—
(10R)-7-amino-12-fluoro-2,10,16-trimethyl-15-oxo-10,15,16,17-tetrahydro-2H-8,4-(metheno)pyrazolo[4,3-h][2,5,11]-benzoxadiazacyclotetradecine-3-carbonitrile.
A name that probably not even its inventors could remember.
Fortunately, it would later receive a much friendlier one – Lorlatinib.
Our Lorla had been born.
────────────────────────
Before any anticancer drug can be tested in humans, it first has to prove itself in the laboratory.
One of the most important parameters is the IC₅₀ value.
It describes the drug concentration required to inhibit cell growth by 50% in vitro.
The lower the IC₅₀ value, the more potent the compound is against its intended target. In the case of PF-06463922, these targets included wild-type ALK, multiple ALK resistance mutations, wild-type ROS1 and several ROS1 resistance mutations.
Among the reported IC₅₀ values were:
• ALK wild type: 2.3 nmol/L
• ROS1 wild type: 0.7 nmol/L
• ALK G1202R: 49.9 nmol/L
• ROS1 G2032R: 196.6 nmol/L
These numbers already told an important story.
First, lorlatinib displayed remarkable potency against both wild-type ALK and ROS1.
Second, they revealed that the ALK G1202R mutation
– and even more so the ROS1 G2032R mutation –
would represent a far greater therapeutic challenge.
────────────────────────
Excellent IC₅₀ values alone, however, do not determine whether a patient will benefit from a drug. What ultimately matters are the concentrations achieved inside the patient’s body.
For Lorla two compartments are particularly important:
• the plasma, reflecting systemic drug exposure, and
• the cerebrospinal fluid (CSF), reflecting drug penetration into the central nervous system.
Both concentrations must be high enough to suppress tumour growth while remaining low enough to avoid unacceptable toxicity.
That sounds straightforward. Unfortunately, biology rarely is.
Two patients taking exactly the same tablet will almost never achieve identical plasma concentrations.
People absorb drugs differently.
They distribute them differently.
They metabolise them differently.
And because the blood-brain barrier also varies between individuals, CSF concentrations differ as well.
In simple terms:
The same tablet can produce very different drug levels in different patients.
For this reason, Pfizer needed to know long before the first patient entered a clinical trial exactly which plasma and CSF concentrations would be required for effective tumour inhibition.
────────────────────────
Once the medicinal chemists had finished their work, it was time for the pharmacologists.
Before the first human volunteer could receive lorlatinib, extensive preclinical studies had to be performed.
As had previously been the case for crizotinib, Pfizer relied on pharmacokinetic/pharmacodynamic (PK/PD) modelling to estimate both a safe starting dose for humans and the plasma and CSF concentrations required for effective ALK inhibition.
Fig. 3 illustrates this process in simplified form.
Experimental laboratory data, animal studies and mathematical PK/PD models were combined to predict the appropriate first-in-human dose.
____________________________________
[Fig. 3 – PK/PD modelling]
____________________________________
Based on toxicology studies in Sprague-Dawley rats and Beagle dogs, together with the PK/PD model, Pfizer selected a starting dose of 10 mg once daily (QD).
The models also predicted the minimum plasma concentrations required to inhibit important ALK resistance mutations:
• ALK wild type : 7.6 ng/mL
• L1196M : 62 ng/mL
• G1202R : 150 ng/mL
Corresponding target concentrations were also estimated for the cerebrospinal fluid:
• ALK wild type : 2.6 ng/mL
• L1196M : 21 ng/mL
• G1202R : 51 ng/mL
These numbers will become surprisingly important later in Lorla’s story.
────────────────────────
You may have noticed that only ALK mutations appear in these tables.
What about ROS1?
Unfortunately, those data have never been published in comparable detail.
Most of the underlying pharmacological analyses remain available only to specialists.
However, given the considerably higher IC₅₀ values observed for ROS1 G2032R, it seems highly likely that substantially higher drug concentrations are required than for the corresponding ALK resistance mutation G1202R.
Would that eventually have clinical consequences?
Be patient. We’ll come back to that.
____________________________________
By the end of the preclinical programme, one thing had become clear.
Lorla worked in the laboratory. It worked in mice. It worked in dogs.
But would it work in humans?
And perhaps even more importantly – at what dose?
That is precisely the purpose of dose-escalation studies.
Because discovering a promising drug is only half the journey.
Finding the right dose is the other half.
And that brings us to the next chapters.
References:
[1] Johnson, T.W. et al. Discovery of (10R)-7-amino-12-fluoro-2,10,16-trimethyl-15-oxo-10,15,16,17-tetrahydro-2H-8,4-(metheno)pyrazolo[4,3-h][2,5,11]-benzoxadiazacyclotetradecine-3-carbonitrile (PF-06463922), a macrocyclic inhibitor of anaplastic lymphoma kinase (ALK) and c-ROS oncogene 1 (ROS1) with preclinical brain exposure and broad-spectrum potency against ALK-resistant mutations. J. Med. Chem. 2014.
[2] Zou HY et al. PF-06463922, an ALK/ROS1 Inhibitor, Overcomes Resistance to First and Second Generation ALK Inhibitors in Preclinical Models. Cancer Cell. 2015
[3] Lin JJ, Choudhury NJ, Yoda S et al. Spectrum of Mechanisms of Resistance to Crizotinib and Lorlatinib in ROS1Fusion-Positive Lung Cancer. Clin Cancer Res. 2021
[4] Jóri B, Falk M, Hövel I, Weist P, Tiemann M, Heukamp LC, Griesinger F. Acquired G2032R Resistance Mutation in ROS1 to Lorlatinib Therapy Detected with Liquid Biopsy. Curr Oncol. 2022
[5] Shaw AT, Chen J et al. Lorlatinib in non-small-cell lung cancer with ALK or ROS1 rearrangement: an international, multicentre, open-label, single-arm first-in-man phase 1 trial. Lancet Oncol. 2017
[6] Chen J et al. Pharmacokinetics of Lorlatinib After Single and Multiple Dosing in Patients with Anaplastic Lymphoma Kinase(ALK)-Positive Non-Small Cell Lung Cancer: Results from a Global Phase I/II Study. Springer NaturE Link. May 2021